1. Field of the Invention
The present invention relates to a focus detecting device for use in a camera or the like.
2. Related Background Art
There is already known a focus detecting device disclosed in the U.S. patent application Ser. No. 457,408 filed Dec. 26, 1989 by the present applicant. Said device is based on a principle of forming a pair of images of an object on a photoelectric converting device from a pair of light beams passing through a phototaking optical system by means of a focus detecting optical system of so-called split pupil type, obtaining image signals of said object by photoelectric conversion of said images by said photoelectric converting device, and effecting a predetermined calculation on said image signals thereby determining the defocus amount of the phototaking optical system.
However, in such focus detecting device of split pupil system, a certain combination of the phototaking optical system and the focus detecting optical system gives rise to a vignetting of the light beams used for focus detection, and, such vignetting, if not uniform, undesirably affects the precision of focus detection and may render the focus detecting operation totally impossible in the worst case.
In the following there will be explained the cause of such vignetting, with reference to FIG. 1.
FIG. 1 illustrates a focus detecting optical system of so-called split pupil system, disclosed in the U.S. patent application Ser. No. 457,408 of the present applicant. Said optical system is composed of a field mask 300 positioned at the primary image plane of the phototaking optical system and provided with an aperture 300A for defining the area of focus detection in the object field, a field lend 301 positioned behind said field mask; re-imaging lenses 303A, 303B, 303C, 303D arranged in two pairs for re-imaging the images of an object on a secondary image plane; and a diaphragm mask 302 positioned in front of said re-imaging lenses and provided with four apertures 302A, 302B, 302C, 302D for limiting the light beams entering said re-imaging lenses. Thus re-focused object images are respectively projected onto photosensor units (for example CCD image sensors) 304A, 304B, 304C, 304D of the photoelectric converting device 304 positioned at the secondary image plane, whereby object image signals corresponding to the light intensity distributions of the object images are generated from said photosensor units.
In the above-explained structure, the forms of four apertures 302A, 302B, 302C, 302D are projected by the field lens 301 onto a pupil plane 305 (hereinafter also called a focus detecting pupil) positioned at a predetermined distance d0 from the primary image plane, wherein the projected forms respectively constitute pupil areas 305A, 305B, 305C, 305D (hereinafter also called focus detecting pupil diaphragms). Consequently the object images re-focused on the secondary image plane are solely formed by the light beams passing through said pupil areas 305A, 305B, 305C and 305D.
FIG. 2 is a cross-sectional view, along a plane containing the X- and Z-axes, of the focus detecting optical system shown in FIG. 1.
The rays passing through the pupil area 305A or 305B and concentrated between end points A and C in the X-direction of the aperture 300A of the field mask 300 have always to pass through an area between hatched portions (said area being defined inside lines A-F and C-H from the primary image plane to the pupil plane 305 at a distance d0 there-from, and inside extensions of lines C-F and A-H beyond said pupil plane 305, wherein F and H are external end points in the X-direction of the pupil areas 305A, 305B). Consequently if the phototaking optical system has a relatively small F-number so that the external end points of the exit pupil thereof are positioned in the hatched portions, the light beams used for focus detection do not cause vignetting and do not have undesirable effect on the focus detection. However, if the F-number becomes larger so that the external end points of the exit pupil are positioned in the internal area, said light beams show vignetting which undesirably affects the focus detection.
The influence of vignetting on the focus detection depends also on the pupil position as well as the size of the exit pupil 101. For example, if the apertures 302A, 302B of the diaphragm mask 302 are shaped like cat's eyes, the pupil areas 305A, 305B on the pupil plane 305 are shaped as shown in FIG. 3. Consequently, the rays which are concentrated on all the points of the aperture 300A of the field mask 300 and then pass through the apertures 302A, 302B should have passed the pupil areas 305A, 305B on the pupil plane 305. Thus, if the exit pupil 101 of the phototaking optical system is positioned at the pupil plane 305, the rays passing through the different points of the aperture 300A and falling onto the photoelectric converting device 304 are subjected to uniform vignetting even in the presence of vignetting. Therefore the vignetting does not cause any undesirable influence but merely results in a uniform loss in the light intensity received by the photosensor units of the photoelectric converting device 304.
However, in any other plane than the pupil plane 305, the rays passing through the different points of the aperture 300A of the field mask 300 and falling onto the photoelectric converting device 304 pass through spatially different areas. For example, light beams which are concentrated on points A, B and C of the aperture 300A shown in FIG. 1 and then pass through the aperture 302B of the diaphragm mask 302, pass through respectively different areas 306A, 306B and 306C shown in FIG. 4 on a plane at a position d1 different from that of the pupil plane 305. Consequently, in case of vignetting when the exit pupil 101 of the phototaking optical system is not positioned on the pupil plane 105, the amounts of vignetting on the light beams passing through different points of the aperture 300A and falling on the photoelectric converting device 304 are not uniform, thus resulting in different losses of light intensity depending on the locations on the photosensor units of the photoelectric converting device 304 and undesirably affecting the focus detection.
FIGS. 5A, 5B, 5C and 5D illustrate the states of such vignetting. In FIGS. 5B, 5C and 5D, points A', B', C', D', E', A", B", C", D" and E" correspond to the points A, B, C, D and E of the aperture 300A shown in FIG. 5A after re-focusing by the lenses 303A, 303B.
FIGS. 5B and 5C show the vignetting in the image of the aperture 300A re-focused by the lenses 303A, 303B. For an object with uniform luminocity, the vignetting does not appear in an area inside the solid line (an area including the point A for the image formed by the re-focusing lens 303A; and an area including the point C for the image formed by the lens 303B), so that the amount of light is 100%. However the vignetting occurs in such a way that the amount of light is reduced to 90-100% in an area between the solid line and the broken line, and is reduced below 90% outside the broken line. When the images formed by the lenses 303A, 303B are superposed, the vignetting-free areas do not overlap each other as shown in FIG. 5D, and exact focus detection becomes impossible because two images to be compared for the focus detection do not coincide each other.
For the above-mentioned drawback, there have been proposed following countermeasures.
For example, in a focus detecting device disclosed in the Japanese Laid-open Patent Sho 55-111927, the above-mentioned drawback is resolved by the use of two focus detecting systems with different F-numbers and by selecting a focus detecting system free from the vignetting, according to the F-number of the phototaking optical system. The F-number for focus detection is determined by the size of a circumscribed circle including the pupil areas 305A, 305B, 305C and 305D on the pupil plane 305 shown in FIG. 1 and the position d0 of said pupil plane.
Also in a focus detecting device disclosed in the U.S. Pat. No. 4,687,917 of the present applicant, the above-explained drawback is resolved by detecting the vignetting state of the focus detecting light beams from the output status of the focus detecting photoelectric converting device, and, in the presence of vignetting, eliminating the low frequency components from the output of the photoelectric converting device, thereby alleviating the influence of said vignetting prior to the calculation for focus detection.
Also in a focus detecting device disclosed in the U.S. Pat. No. 4,816,663 of the present applicant, the above-mentioned drawback is resolved by calculating the amount of vignetting from the F-number of the exit pupil of the phototaking optical system and the positional information of said exit pupil, and effecting the calculation for focus detection so as to reduce the influence of the vignetting according to thus calculated amount of vignetting.
However, such conventional focus detecting devices have been associated with considerable errors in the calculation of loss in the amount of light, because the losses in peripheral light in the phototaking optical system and in the focus detecting optical system are not included in the loss of light resulting from vignetting. For this reason switching of focus detecting calculation process according to the loss of light amount cannot be conducted in exact manner. It has also been difficult to use the output of photoelectric conversion for the focus detection after the correction for loss of light amount, or to use a portion of said output free from loss in light amount. This drawback has become more serious with the recent expansion of the focus detecting area with respect to the photographing image area, as the losses in the peripheral light amount in the phototaking optical system and in the focus detecting optical system are no longer negligible.